Schedule

Check-In & Breakfast

Coffee & Pastries will be provided

Opening Remarks 

Gaurav Sukhatme, Director of the School of Advanced Computing, Donald M. Alstadt Chair in Advanced Computing, Professor of Computer Science, University of Southern California

Envisioning the Next Era of AI Research at USC: A Roadmap to Societal and Scientific Impact

Gaurav Sukhatme, Director of the School of Advanced Computing, Donald M. Alstadt Chair in Advanced Computing, Professor of Computer Science, University of Southern California

Shrikanth (Shri) Narayanan, Vice President for Presidential Initiatives, Research Director of the Informational Sciences Institute, University Professor, Niki & C. L. Max Nikias Chair in Engineering, University of Southern California

John Blosnich, Director of the Center for LGBTQ+ Health Equity, Interim Co-Director of the USC Center for AI in Society, Associate Professor at the Suzanne Dworak-Peck School of Social Work, University of Southern California

Michàlle Mor Barak, Dean Endowed Professor at the Suzanne Dworak-Peck School of Social Work and the Marshall School of Business, University of Southern California

Health

DermEquity: Auditing and Mitigating Skin Tone Bias in Dermatological AI Systems - Parth Gosar, Viterbi College of Engineering, University of Southern California

Navigating Digital Mental Health: Predictors and Barriers to DMHI Adoption Among AAPI Emerging Adults - John Bunyi, CAIS Student Leader, Suzanne Dworak-Peck School of Social Work, University of Southern California

Evaluating Data Modalities, Model Scale, and Safety in On-Device Stress Prediction - Ibukunoluwa Soyebo, Viterbi College of Engineering, University of Southern California

Local-to-Global (L2G) Intervention Design via Counterfactual Explanations on Graphs - A Case Study in Suicide Risk Prevention - Mulin Tian, Viterbi College of Engineering, University of Southern California

Spatiotemporal Social Media Sentiment as a Predictor of Real-Time Mood - Sean Minns, Dana and David Dornsife College of Letters, Arts and Sciences, University of Southern California

The Voices in the Loop (VITL) Study: Participatory Systems Modeling for Suicide Prevention with LGBTQ+ Young Adults Experiencing Homelessness - Maiya Hotchkiss,  Suzanne Dworak-Peck School of Social Work, University of Southern California

Conversational AI for HPV Vaccination: Mitigating Provider Burnout and Improving Patient Counseling - Sadra Sabouri, Viterbi College of Engineering, University of Southern California; Helen Horan, Keck School of Medicine, University of Southern California; Xiaoyue (Mona) Guou, Keck School of Medicine, University of Southern California; Kartik Vasudev Shenoy

Privacy-Preserving and Sustainable Biomechanical Modeling from Wearable Sensors for Health, Wellbeing, and Assistive Systems - Bowen Song, Viterbi College of Engineering, University of Southern California

Brain Tumor Detection in MRI Images - Khushi Kumari, Viterbi College of Engineering, University of Southern California

Opportunities and Risks of Artificial Intelligence in Mobile Health: A Patient-Centered Perspective - Alyssa Donawa, Viterbi College of Engineering, University of Southern California

Modeling Community Attitude through Reaction Tone: A Human-AI Collaborative Framework for Evaluating LLM Alignment with Community-Situated Linguistic Behaviors - Nuan Wen, Viterbi College of Engineering, University of Southern California

Environment

Towards Trustworthy Vision-Language Systems for Air Pollution Monitoring: A Knowledge Graph Driven Architecture - Kaushal Patil, Viterbi College of Engineering, University of Southern California

Teaching AI-Enabled Social Media for Social Impact: An Experiential, Project-Based Model for Sustainability and Behavior Change - Misha Kouzeh,  Annenberg School for Communication and Journalism, University of Southern California

An Equivariant MARL Algorithm for Sensor Placement on Dynamic Networks - Thomas Reeves, Viterbi College of Engineering, University of Southern California

Uncovering the Environmental Blind Spot: A Systematic Review of Artificial Intelligence Ethics, Social Work, and Environmental Sustainability - Nick Walker-Craig, Suzanne Dworak-Peck School of Social Work, University of Southern California

Homelessness

Listening to the City: Using AI to Detect Early Warning Signals of Homelessness from Urban Stress Data - Ishita Gopalakrishnan, Viterbi College of Engineering, University of Southern California

Toward Privacy-Preserving Simulation of PEH Survey Populations Using Large Language Models - Brihi Joshi, Viterbi College of Engineering, University of Southern California

AI Methods for Good

Hostility, Solidarity, and Constructive Dialogue in Online Discussion on Abortion Rights - Lei Cao, Annenberg School for Communication and Journalism, University of Southern California

The Trafficker’s Pitch: Detecting Deceptive Recruitment in Online Job Boards - Siyi Zhou, Annenberg School for Communication and Journalism, University of Southern California

Education

Integrating AI to Support Project-Based Learning in K-12 Classrooms - Ahmya Rivera, Iovine and Young Academy, University of Southern California

Social Media

How Far Will They Go? Red Teaming Political Expressions of Large-Language Models - Daniel Ruiz, Viterbi College of Engineering, University of Southern California

Work & Labor

Designers as Co-Creators: Reconfiguring Workflows and Creative Agency in AI-Augmented Design - Runsheng Zhang,  Iovine and Young Academy, University of Southern California

CounselBench: A Large-scale Expert Evaluation and Adversarial Benchmarking of Large Language Models in Mental Health Question Answering

Yahan Li, PhD Student, Viterbi School of Engineering, University of Southern California
John Bosco S. Bunyi, CAIS Student Leader, PhD Student, Suzanne Dworak-Peck School of Social Work, University of Southern California

Interpretable Machine Learning for Early Substance Abuse Risk Prediction in Pediatric Health Care

Vidur Mushran, CAIS++, Undergraduate Student, Dana and David Dornsife College of Letters, Arts and Sciences, University of Southern California
Sonya L Negriff, Kaiser Permanente

Advancing Fairness in Medical Resident Rotation through Integer Programming

Karmel S. Shehadeh, Assistant Professor, Viterbi College of Engineering, University of Southern California 

Forecasting Acute Child Malnutrition in Kenya

Jay Campanell, CAIS++, Undergraduate Student, Dana and David Dornsife College of Letters, Arts and Sciences, University of Southern California
Bistra Dilkina, CAIS Co-Director, Associate Professor, Viterbi School of Engineering, University of Southern California

Q&A

The Subjectivity of Respect in Police Traffic Stops: Modeling Community Perspectives in Body-Worn Camera Footage

Jackson Trager, PhD Student, Dana and David Dornsife College of Letters, Arts and Sciences, University of Southern California

No Free Lunch in Language Model Bias Mitigation? Targeted Bias Reduction Can Exacerbate Unmitigated LLM Biases

Faith Baca, PhD Student, Viterbi College of Engineering, University of Southern California

Decoding the Illicit Wildlife Trade: From Physical Airport Hotspots using AI to Digital Detection using LLMs

Trent Kobielusz, Undergraduate Student, Viterbi School of Engineering, University of Southern California
Jeena Mahajan, Undergraduate Student, Viterbi School of Engineering, University of Southern California
Hannah Murray, CAIS Student Leader, PhD Student, Viterbi School of Engineering, University of Southern California 

Q&A

The Use of AI to Solve the Toughest Challenges Faced by Military and Veteran Populations

Ajitesh Srivastava, Assistant Professor, Viterbi School of Engineering, University of Southern California
Carl Castro, Professor, Director of Military and Veterans Programs, Suzanne Dworak-Peck School of Social Work, University of Southern California

Health

Tied In on TikTok: Tie Strength and Emotional Dynamics in Algorithmic Communities - Charles Bickham, Viterbi College of Engineering, University of Southern California

Overstating Attitudes, Ignoring Networks: Large Language Model Biases in Simulating Misinformation Susceptibility - Eun Cheol Choi,  Annenberg School for Communication and Journalism, University of Southern California

Multi-Turn Interactions Reveal Hidden Safety Failures in Mental Health Language Model Use - Jaspreet Ranjit, CAIS Student Leader, Viterbi College of Engineering, University of Southern California

Social Outcome Prediction and Human Intervention Engine (SOPHIE): A Human-Centric AI Partner for Navigating Complexity in Social Welfare Policy - Stephen Lieberman, Suzanne Dworak-Peck School of Social Work, University of Southern California

Evaluating and Mitigating Socioeconomic Inequities in Patient-Facing Health Information from Large Language Models: The Role of Social Determinants of Health - Shwetha Sadanand, Viterbi College of Engineering, University of Southern California

CliniSyc: Evaluating LLM Sycophancy in Real-World Multimodal Clinical Diagnostic Queries - Thor Christoffersen Hochman & Brihi Joshi, Viterbi College of Engineering, University of Southern California

From Image Bias to Temporal Power: Auditing Social Inequality in AI-Generated Videos - Zituo Wang, Annenberg School for Communication and Journalism, University of Southern California

Contextualizing Harm: Modeling Subjectivity in Automated Toxicity Detection - Naysa Bhargava, Viterbi College of Engineering, University of Southern California

Behavioral Synchrony And Parasympathetic Response In Mother-Infant Play - A'di Dust, Viterbi College of Engineering, University of Southern California

Rethinking Digital Social Support: Online Connection and Employment-Related Mental Health Risk in Young Adulthood - Chaejeong Lee, Suzanne Dworak-Peck School of Social Work, University of Southern California

Skilled Resilience: Revitalizing Asian American and Pacific Islander Entrepreneurship Through AI-Driven Social Media Marketing Techniques - Vanya Shrivastava, Viterbi College of Engineering, University of Southern California

Environment

Machine Learning Guided Optimal Transmission Switching to Mitigate Wildfire Ignition Risk - Weimin Huang, Viterbi College of Engineering, University of Southern California

FireLoc: Real-Time Wildfire Detection and Localization System - Yash Gupta, Viterbi College of Engineering, University of Southern California

Reducing the Carbon Footprint of AI Reasoning: A Meta-Cognitive Approach to Computational Efficiency - Yashaswi Sharma, Viterbi College of Engineering, University of Southern California

Homelessness

Framing deservingness: An analysis of public communication of homelessness services in Los Angeles County - Vivian Zhai, Annenberg School for Communication and Journalism, University of Southern California

Learning Robust Risk Scores Under Unobserved Confounders - Ryan Edmonds, Viterbi College of Engineering, University of Southern California

Empty Stomachs, Restless Nights: Linking Housing Instability, Breakfast Skips, and Sleep Deficits to Academic Outcomes Using 2023 Youth Risk Behavior Survey Data - Geoffrey Cui, Suzanne Dworak-Peck School of Social Work, University of Southern California

AI Methods for Good

Cross-Platform Discourse Networks - Patrick Gerard, Viterbi College of Engineering, University of Southern California

Does Shared Social Identity Promote Social Coordination - Ashwin Rao, Viterbi College of Engineering, University of Southern California

Information suppression in large language models: Auditing, quantifying, and characterizing censorship in DeepSeek - Peiran Qiu, Viterbi College of Engineering, University of Southern California

Public Policy

A Deterministic, MCP-Mediated Evidence Engine for Auditable Municipal Finance Analytics via Local LLM Orchestration - Chirag Shivakumar, Viterbi College of Engineering, University of Southern California

Education

Modeling Fine-Grained Multi-Modal User Disengagement in Screen-Based Educational Activities for Children with Autism Spectrum Disorder - Aakanksha Peeru, Viterbi College of Engineering, University of Southern California

Capturing Behavioral Shifts in Cannabis Dependence Risk Using Transfer Learning across Policy Contexts

Jiyang He, PhD Student, Viterbi School of Engineering, University of Southern California
Shinyi Wu, Professor, Suzanne Dworak-Peck School of Social Work & Viterbi College of Engineering, University of Southern California

AI's Human-Like Tone and Flattery Impact on Clinical Therapy

Sadra Sabouri, PhD Student, Viterbi College of Engineering, University of Southern California
Yalda Daryani, PhD Student, Dana and David Dornsife College of Letters, Arts and Sciences, University of Southern California

Forest-Dependent Species Projections May Overestimate Future Ranges by Overlooking the Synergistic Impacts of Climate, Dispersal, and Wildfire

Pratyush Jaishanker, Undergraduate Student, Viterbi College of Engineering, University of Southern California 

Optimizing Sensor Placement for Wildlife Monitoring

Hannah Murray, CAIS Student Leader, PhD Student, Viterbi School of Engineering, University of Southern California 

Q&A